Inspiration
Knowledge is the foundation for solving all problems. With enough knowledge, any problem that isn’t limited by the laws of physics can be solved.
While knowledge is abundant, the desire to learn is scarce. If we can accelerate learning, we can solve problems faster and get better problems. However, the way we currently approach learning is flawed.
What it does
MindGraph addresses this issue. We help people learn faster by streamlining the learning process. The app searches a topic and presents a skill tree, showing the skills required to master an advanced subject. For example, if you want to learn about nuclear fusion, you can see all the subtopics leading up to it.
When you click on a node in the tree, we use Cohere to search the web for relevant information and provide a quick explanation. Then, we use Claude to generate questions on the topic. If you struggle to answer a question, we implement spaced repetition and let you rate your understanding.
How we built it
We developed MindGraph using Next.js and TypeScript, integrating Cohere, Claude, Cursor, and OpenAI.
Challenges we ran into
Too much coffee.
Accomplishments that we're proud of
We successfully created a visual knowledge tree and designed a user-friendly interface for learning.
What we learned
Think small during hackathons.
What's next for Emil Gang
We plan to develop a social graph that connects people who are on the same learning journey, enabling them to exchange ideas. Imagine seeing others exploring esoteric topics online and making friends that way—a kind of "echo chamber as a service."
It would also be exciting to track who has mastered the most skill trees.
Future features:
- Allow users to add their own resources and break down books into digestible formats.
Built With
- claude
- cohere
- nextjs
- reactflow
- shadcn
- typescript
Log in or sign up for Devpost to join the conversation.